自調式沖淡濾波器之發展及其於虛擬實境船舶動態模擬系統之應用

Abstract

以史都華六軸運動平台(Stewart Platform)為基礎之動感模擬器，在軍事模擬、娛樂、教育訓練等方面皆可見到其應用。動感模擬器之原理是利用移動可動平台，讓駕駛員感受到相當程度的直線和旋轉的速度和加速度，以達到模擬的效果。然而史都華平台的工作空間是有限的，所以我們需要利用沖淡濾波器(Washout Filter)，將人體無法感受的高頻和低頻部份濾除，減少史都華平台的動作量，同時還要保留駕駛時的各種感受；此外，還能在適當的時候將平台帶回到原點附近，以提供較大的工作空間。因此動感模擬器所提供的動作提示(motion cues)是否逼真，沖淡濾波器是一個相當重要的關鍵。 本論文以沖淡濾波器為主體，透過對人體平衡器官及人類運動感覺模型的了解，利用汽車動態模型對沖淡濾波器之功能做驗證，並結合史都華六軸運動平台與實驗室自行發展的船舶動態模型，完成一個完整的虛擬實境船舶動態模擬系統。 在沖淡濾波器設計方面，本論文利用模糊控制器，修正動感模擬器的錯誤動作，並基於適應性控制法則及類神經模糊網路理論，設計具自我調整能力之沖淡濾波器架構。

Motion simulators based on Stewart Platform are used widely in military simulation, entertainment, educational training, etc. In order to make a simulation more realistic, feelings of linear accelerations and angular rates are exerted on the pilot by moving the mobile platform. This has to be accomplished without driving the simulator out of its workspace, and the motion-drive algorithm that is in charge of this is commonly referred to as washout filter. Therefore one of the keys to obtain realistic motion cues on the simulator pilot is the washout filter. The main purpose of this thesis is to achieve a VR-base motion simulation system and to construct a scheme of auto-tuning washout filter. In the design of washout filters, we adopted a fuzzy logic controller to tune the motion errors generated by motion simulators. In addition, by using of adaptive control algorithm and neuro-fuzzy network, we proposed an auto-tuning washout filter that adjusts its parameters automatically to fit the various purposes of applications. In experiments, by the study of human sense organs and the mathematical model of vestibular system, we used a simplified vehicle dynamic model to verify the functions of washout filters. Afterward we integrated the Stewart Platform and a ship dynamic model developed by our laboratory into a complete VR-based ship dynamic simulation system.